INFLAMMATORY DISORDERS ASSOCIATED WITH HELICOBACTER PYLORI IN THE ROUX-EN-Y BYPASS GASTRIC POUCH

ABSTRACT Background: The prevalence of Helicobacter pylori in obese candidates for bariatric surgery and its role in the emergence of inflammatory lesions after surgery has not been well established. Aim: To identify the incidence of inflammatory lesions in the stomach after bariatric surgery and to correlate it with H. pylori infection. Methods: This is a prospective study with 216 patients undergoing Roux-en-Y gastric bypass. These patients underwent histopathological endoscopy to detect H. pylori prior to surgery. Positive cases were treated with antibiotics and a proton inhibitor pump followed by endoscopic follow-up in the 6th and 12th month after surgery. Results: Most patients were female (68.1%), with grade III obesity (92.4%). Preoperative endoscopy revealed gastritis in 96.8%, with H. pylori infection in 40.7% (88/216). A biopsy was carried out in 151 patients, revealing H. pylori in 60/151, related to signs of inflammation in 90% (54/60). In the 6th and 12th month after surgery, the endoscopy and the histopathological exam showed a normal gastric pouch in 84% of patients and the incidence of H. pylori was 11% and 16%, respectively. The presence of inflammation was related to H. pylori infection (p<0,001). Conclusion: H. pylori has a similar prevalence in both obese patients scheduled to undergo bariatric surgery and the general population. There is a low incidence of it in the 6th and 12th months after surgery, probably owing to its eradication when detected prior to surgery. When inflammatory disease is present in the new gastric reservoir it is directly related to H. pylori infection.


H
. pylori infection has an incidence of 24-67% among bariatric patients. Upper gastrointestinal endoscopy (UGE) is used prior to surgery to detect this bacteria, in view of its high incidence and possible relation with pathological abnormalities of the stomach. In some locations, such as Finland, UGE isprerequisite for all bariatric patients, although this practice is still questioned 9,14 .
Inflammatory diseases of the stomach after bariatric surgery, especially Roux-en-Y gastric bypass (RYGB), include gastritis and ulcers (of the new gastric reservoir and the anastomosis). There is no difference in the etiopathogeny of these lesions in the operated or non-operated stomach, with H. pylori being the main cause and non-steroid antiinflammatory drugs the secondary cause. However, the relation between these lesions and RYGB is not fully understood 9,11 . UGE to detect H. pylori prior to bariatric surgery has been required in triage for the presence of this bacteria by health insurance plans, especially in cases of RYGB. This requirement is based on the supposition that the existence of these bacteria is linked to ulcers or cancers of the excluded stomach after the procedure. The present study uses a number of tests to conclude its diagnosis, including the rapid urease test, histology and tissue biopsy, along with non-endoscopic tests of blood and serum 10,13 .
The aim of this study was to identify the incidence of inflammatory lesions in the stomach after bariatric surgery and to correlate it with H. pylori infection.

METHODS
The study was approved by the Research Ethics Committee of the Federal University of Pará (Tropical Medicine Unit), Belém, PA, Brazil. All patients were studied in accordance with the precepts of the Helsinki Declaration and the Nuremberg Code and the norms for research involving human beings were respected (Res. CNS 196/96).
A prospective study was carried out with two groups of patients from the Bariatric Surgery Service of Hospital Porto Dias in Belém, PA, Brazil. The two groups underwent surgical treatment for obesity, in accordance with CFM Resolution No. 1,766/05.
The first group was used to study the prevalence of H. pylori infection in obese patients through histopathological examination of fragments obtained by endoscopic biopsy prior to surgery. Patients testing positive underwent eradication treatment using antibiotics as outlined in the 2 nd Brazilian Consensus on the Study of H. pylori, using a combination of PPI, clarithromycin and amoxicillin.
The second group was composed of at least 100 patients of the first group, who underwent endoscopy at 6 and 12 months after surgery, in order to evaluate the incidence of bacteria and inflammatory diseases of gastric pouch.

RESULTS
In the first phase of the study, 2010-2012, 216 obese patients indicated for surgery were analyzed. 147/216 (68.1%) were female and 69/216 (31.9%) male; most patients were in the third or fourth decade of life, 69 (31.9%) and 67 (31%) respectively, with progressively fewer in the older age groups and few in the second decade.
The prevalence of H. pylori by gender was similar for both sexes, 28/88 (40.6%) for men and 60/88 (40.8%) for women and there was not statistically significant correlation (p=0.9736). The distribution by age group and BMI showed a difference although this was not statistically significant (p<0.3114).
Analysis of the presence of inflammatory activity in the gastric mucosa prior to surgery was carried out in 151 of the 216 patients studied. Of these 60/151 tested positive for H. pylori and 54/60 (90%) had a histologically active inflammatory process, compared to 26/91 (28.6%) of patients in whom the bacteria was not found, and this was a significant difference (p<0.001). The likelihood of the presence of H. pylori among patients with inflammatory activity was 22 times greater than in patients without such activity (OR=22.5, Table 1). The test for H. pylori found 13 (11.9%) of these 109 patients to be positive and 96 (88.1%) negative.
Among the 92 patients with normal endoscopy six months after surgery, the incidence of H. pylori was 7 (7.6%), while H. pylori was present in 6 (35.3%) of the 17 patients with endoscopic gastritis (p<0.0047). The likelihood of H. pylori being present in the patients with gastritis was six times greater than among patients without gastritis (OR=6, Table 2). 109 Histological analysis after six months was carried out in 54 patients to investigate the presence of inflammatory activity in the gastric mucosa in this group and H. pylori was found to be present in all nine patients with this activity (100%, p<0.0001, Figure 1). In the group of patients evaluated 12 months after surgery, 125 underwent UGE, of whom 105 (84%) presented with a normal endoscopy, 15 (12%) with gastritis and 5 (4%) with an ulcer in the new reservoir. Including gastritis and ulcers, there were 20 (16%) cases of inflammatory disease in the operated stomach.

TABLE2 -Correlation between results of endoscopy and H.pylori infection six months after surgery
Of these 125 patients, 19 (15.2%) tested positive for H. pylori and 106 (84.8%) negative (Table 3). Of the 105 patients with normal endoscopy 12 months after surgery, H. pylori was found in 14 (13.33%), while, in the OriginAl Article 20 patients with endoscopic gastritis, H. pylori was present in 5 (25%, p<0.3211, Table 4). Analysis of the correlation between H. pylori and inflammation of the gastric mucosa showed that 17 of the 59 patients undergoing endoscopic biopsy (28.81%) had inflammatory activity and nine of these tested positive for H. pylori, compared to two among the 40 normal examinations (p<0.0001). the likelihood of H. pylori being present in patients with histological inflammatory activity was 22 times greater than in patients without this activity (OR=22.5) ( Table 5). There were no statistically significant alterations in the incidence of H. pylori for the variables age, gender, or BMI.

DISCUSSION
There are divergences in the literature as to the prevalence of H. pylori in the obese. In Saudi Arabia, it is found in 68-82.2% of the population and is attributed to socioeconomic and sanitary factors. In the obese, the bacteria was present in 85.5% of patients who have undergone bariatric surgery 1 .
A systematic review has shown that the prevalence of H. pylori in obese patients scheduled to undergo bariatric surgery varies from 6.9-61.3%. The prevalence of infection caused by this pathogen varies from 30 to 90% around an average of 60% 7 .
In a national study, the prevalence of H. pylori was 60%. The authors recommended the use of two methods to research the bacteria (urease and histology) to increase accuracy 2 .
RYBG surgery involves resection a part of the stomach that is called the excluded stomach. This stomach has a high probability of developing abnormalities that may be the consequence of bile and pancreatic secretion reflux. H. pylori may be one of the causes of some dysfunctions and should be treated with caution prior to surgery, since the exclusion of this part of the stomach makes access to it difficult 16 .
H. pylori infection causes inflammation of the gastric mucosa and may lead to problems such as intestinal metaplasia and even cancer. Its eradication may revert this inflammatory process but this is not possible in more advanced phases 7 .
The need for endoscopy prior to surgery is still controversial. In a study conducted by Wong et al. with 180 patients undergoing gastric bypass, an alarming number of 159 were diagnosed with chronic superficial gastritis and esophageal reflux, erosion, hiatal hernia and gastric ulcer were also found in smaller numbers of patients 18 .
In a recent literature review Palermo et al. showed that the presence of H. pylori prior to surgery may be related to the development of postoperative marginal ulceration. Thus, patients with upper gastrointestinal symptoms should undergo endoscopy prior to gastric bypass and be treated for H. pylori if they test positive. However, some authors believe that the prevalence in patients undergoing RYGB is similar to that of the general patient and that the H. pylori test and preoperative treatment do not diminish the incidence of anastomotic ulcer or gastritis in the gastric pouch 12 .
Apart from UGE, a biopsy is also fundamental in determining the future management of the surgical procedure and may shift it to initial treatment of an existing pathological abnormality. H. pylori is already known to be a carcinogenic agent, which operates by way of chronic gastritis or intestinal metaplasia. These changes in the stomach undergoing RYGB surgery may be harmful, because of the existence of the excluded stomach, leading to serious complications, if abnormalities are not identified prior to the procedure 6,7 .
Considering the possible endoscopic alterations found in the UGE on patients undergoing bariatric surgery, research suggests a classification of endoscopic findings in the preoperative RYGB, reinforcing the importance of preoperative screening 5 .
In the present study of 216 patients, the prevalence of H. pylori was 40.7% and there were no statistically significant differences in terms of sex, age group or BMI. Histopathological analysis of the mucosa proved to be significant (p<0.001) with the bacteria responsible for inflammatory activity in 90%. In a study of 854 patients undergoing bariatric surgery, the prevalence of H. pylori was around 23.7%, but the article cites other sources giving a range of prevalence of the bacteria that varies from 11.5-66.7% 17 .
The present study eradicated H. pylori in patients who had tested positive for the bacteria prior to surgery. Treatment followed the schema outlined in the 3 rd Brazilian Helicobacter pylori Consensus, with an eradication rate of nearly 80% 4,8 . The 109 patients who had undergone RYGB six months earlier were symptomless and 84.4% of these presented with a normal endoscopy and 15.6% with inflammatory disease of the new reservoir. The incidence of H. pylori in patients was 11.9%, but the incidence in those with inflammatory disease of the new reservoir was 35.3% (p<0.004).
The likelihood of testing positive for H. pylori in patients with inflammatory disease of the new reservoir is six times greater than in those without inflammatory disease, strongly indicating a relation between the presence of the bacteria and inflammatory lesions of the operated stomach. However, there is controversy in the literature regarding the presence of this pathogen and inflammatory lesions. Rawlins et al. showed, in 228 patients undergoing RYGB, that there was no evidence of a connection between H. pylori and an increase in the postoperative complications rate, further underlining the importance of this study in scientific circles 15 .
This becomes even more apparent when the histopathological exams of these patients are taken into consideration. Crosstabulation of the presence of inflammatory activity of the gastric mucosa with the presence of H. pylori, showed that nine of the 53 exams conducted revealed active inflammatory activity and all showed infection with H. pylori (p< 0.001).
In the 12 th month after surgery, 125 symptomless patients were evaluated and 84% had normal endoscopy, while 16% had inflammatory disease of the new reservoir. H. pylori was present in 15.2%, a little higher than the incidence in the 6 th month after surgery but without statistical significance (p<0.3211).
Analysis of the gastric mucosa of 59 patients after 12 months revealed 17 with inflammatory activity, nine of whom tested positive for H. pylori, compared to two of the 42 histopathological exams with absence of inflammatory activity (p<0.001). The likelihood of H. pylori being present in patients with inflammatory activity was 22 times greater than in patients without such activity, clearly indicating the relation between inflammatory disease of the new gastric reservoir and H. pylori infection.
In the patients studied, the low incidence of ulceration inFlAMMAtOrY DiSOrDerS ASSOciAteD WitH HelicOBActer PYlOri in tHe rOUX-en-Y BYPASS gAStric POUcH of the gastric stump and gastritis may be related to routine eradication of H. pylori in our protocol. Furthermore, other studies have shown that eradication of H. pylori may be related to a decrease in the incidence of perforations of the viscera and postoperative marginal ulcers. In one study of 560 patients, the incidence of ulceration was 2.4% in tested and treated patients, compared to 6.8% in another study where this protocol was not applied 3 .
The study did not include a test after treatment to confirm eradication of H. pylori in the 6 th and 12 th month after surgery, given the failure rate of around 10% for the classical treatment. It is thus important to note that patients testing positive for H. pylori after surgery need to have their data cross-tabulated with preoperative data to evaluate whether they were already positive and, in this case, to opt for second line treatment, thereby avoiding failure for reason of bacterial resistance.
Research into both H. pylori and possible lesions of the excluded stomach poses a challenge for scientific studies, owing to the possible emergence of overwhelming technical difficulties, sometimes making it impossible to conduct the procedure. The difficulty is not restricted to research but also impedes treatment 7 .
The results obtained by the present study indicate the importance of diagnosis of the presence of H. pylori in patients undergoing bariatric surgery, especially when the RYGB technique is used, since this technique involves excluding part of the stomach, which may lead to the emergence of inflammatory diseases in the new gastric reservoir.

CONCLUSIONS
H. pylori has a similar prevalence in both obese patients scheduled to undergo bariatric surgery and the general population. There is a low incidence of it in the 6 th and 12 th months after surgery, probably owing to its eradication when detected prior to surgery. When inflammatory disease is present in the new gastric reservoir it is directly related to H. pylori infection.